A Fast and Accurate Model of Thoracic Disease Detection by Integrating Attention Mechanism to a Lightweight Convolutional Neural Network

The utilization of Deep Learning, especially the Convolutional Neural Network (CNN), is currently the best approach for thoracic disease detection from Chest X-Ray images. However, CNN typically has a slow runtime, hence potentially can introduce a bottleneck in the healthcare that uses the technology. To answer the challenge, we proposed a model that integrates an attention mechanism to a lightweight CNN. The proposed model can run faster than the current state-of-the-art model for thoracic disease detection while having the second-best performance among the thoracic disease detection models.

International Conference on Computer Science and Computational Intelligence 2020
Branden Adam Sangerokia and Tjeng Wawan Cenggoro